Public Accounts Commission
Oral evidence: National Audit Office Audit Quality 2021-22
Tuesday 6 December 2022
Ordered by the House of Commons to be published on 6 December 2022.
Members present: Mr Richard Bacon (Chair); Jack Brereton; Anthony Browne; Peter Grant; Dame Meg Hillier; Jerome Mayhew.
Questions 1-60
Witnesses
I: Gareth Davies, Comptroller & Auditor General, National Audit Office; Janet Eilbeck, Independent Chair, Audit Quality Board, National Audit Office; Elaine Lewis, Executive Director, National Audit Office; Kate Mathers, Executive Director, National Audit Office.
Witnesses: Gareth Davies, Janet Eilbeck, Elaine Lewis and Kate Mathers.
Q1 Chair: Welcome to this meeting of the Public Accounts Commission on Tuesday 6 December. We are joined by the Comptroller and Auditor General, Gareth Davies; Kate Mathers, the executive director with responsibility for the financial audit service line at the NAO; Elaine Lewis, the executive director with responsibility for financial audit quality; and Janet Eilbeck, who is a non-executive director of the National Audit Office and the independent chair of the audit quality board of the NAO. You are all very welcome. Before we start our questions, I invite Peter Grant to make a declaration on the public record.
Peter Grant: Thank you, Chair. Can I place again on the public record that I am a member of one of the major professional accountancy and audit organisations, which supplies a steady stream of staff to the National Audit Office?
Chair: Thank you very much. May I start, C&AG, with you? We are looking at this review of audit quality. According to the Financial Reporting Council, nearly half of the NAO audits that it reviewed fell below the quality threshold. That follows a trend of declining audit quality scores in recent years. How concerned are you about that?
Gareth Davies: Well, the scores for the 2021 audits, which are the most recent ones looked at, are not where they should be, so we are obviously very concerned about that, and we will explain what we have been doing and will continue to do to improve quality. On your point about the trend, for the last two years our scores have been roughly where the leading firms’ scores have been. In the most recent year, we have fallen below that pack, which is the point of concern. Restoring us to the position where we were in the last two years will be an important first step. Of course, we want to push on beyond that. At the equivalent hearing last year, I said that our target is to have all our audits in the top two categories used by the FRC. That remains the case, and we have a detailed programme of work to achieve that.
Q2 Chair: One of the things that the Financial Reporting Council points out is the need for you to consider both the effectiveness and the timeliness of your previous actions to address audit quality. To what extent could it be said that your previous actions have fallen short?
Gareth Davies: I don’t think they have, actually. The really important point to be clear on as we discuss these results is the time lag involved. These are the audits of the 2021 Government accounts, which we started planning two years ago in the autumn of 2020. We delivered the work in the summer and autumn of 2021. In order to show through in the audits being inspected here, any improvements would need to have been the result of actions that we took before the autumn of 2020, two years ago. Everything that we have done since then, which is a lot, is not reflected in those audits and these results. The frustrating thing for us and for you is that we have to wait two years for the full impact of our improvement actions to show in the inspection results, given the retrospective nature of audit and the inspection of audit.
Q3 Chair: This Commission has previously discussed staff turnover, retention, and workload at the National Audit Office. Do you think that those things have, to some extent, affected audit quality, and that that has shown through in your audit quality review results?
Gareth Davies: No, again partly because of the lag effect. The pressure on workforce, both for us and for the firms, has been more recent than would have shown up in these results. For the reason I gave before, I would not attribute it to that. Clearly, the audits covered here—the 2021 audits—were the first audits affected by the pandemic. First, there is the very good point that none of the FRC’s findings expressed any concern about the impact of, say, remote working on the quality of our work. It found no evidence that the enforced move to remote working for these audits had had any impact on quality, but clearly, as we said in our note to the Commission, from 2020-21 and 2021-22 those audits were heavily affected by the pandemic spending and the impact of very high-risk schemes. It was not one of the ones selected for review, but the Department of Health and Social Care audit for 2021 is more than double the size of the same audit the previous year, so there was a lot going on in these audits. However, I do not think that that is the explanation for the findings.
As the FRC says, the key issue that it picked up on in the audits that it judged to require improvement was our reliance on other experts. That relates to quite a wide range of things. For example, when we audit a public body with a large pension scheme, we rely on the audit of the pension scheme by a firm, or on the valuation of assets, including investments, by fund managers and other experts. Of course, auditing standards require us to get our own expert advice in these situations. Essentially, we are being challenged on the extent to which that goes far enough and deep enough. There is a question of degree here, rather than whether we know the right things to do. I think that the findings say that we do; what we are being challenged on is how far we take that work. Of course, that is essentially a rising bar for the profession as a whole. We are, as with the rest of the profession, catching up with expectations about how much of that is necessary.
Q4 Chair: How do you incentivise quality among your staff? I am sure that the audit transformation programme will make a number of process-based changes, which will hopefully make things better, but does it seek to make cultural change as well?
Gareth Davies: Yes. Culture change is the bedrock of our quality improvement plan. We will go on to explain why that is. It is No. 1 in our organisational priorities. It is the No. 1 on our risk register that we disclose in our annual report and accounts. Nobody is under any illusions about the importance of this in the NAO. The culture that we have worked really hard to create in the last two or three years is one of a commitment to improvement, and openness about the findings, including shortcomings, of our internal reviews and the FRC’s reviews. We have an open discussion about those, and obviously put plans in place to address them.
That culture is very important, because what I found at the NAO was a bit of a fear culture around this. If you got a poor score on one of your audits, you definitely did not talk about it, and you hoped that nobody else would ask you about it. We made it quite difficult to have open conversations about that. That has changed completely. We do not celebrate poor scores—far from it—but we learn from them and expose the lessons, to make sure that we do not waste the opportunity to improve in those areas for the future. That bit of culture is very important.
On your question about how we incentivise quality, in running the NAO, we operate a balanced scorecard. There are essentially four quadrants that we monitor every month. Quality is one of those four, along with delivery, people and resources. That is how every senior leader in the NAO, including our audit directors, is judged in their performance review. It is how I am judged by the board each year for our performance as a whole organisation. It is clear. We do not have too many priorities, so it does not get lost in a long list of priorities, and you cannot get a positive performance review as an individual auditor without tackling audit quality. We ensure that there is an audit quality objective in everybody’s set of annual objectives.
Q5 Chair: One of the things that we have discussed previously is your difficulty in recruiting trainees. You recruited fewer than you hoped to recently. Presumably that places extra pressure on those leading audits. How do you mitigate that risk?
Gareth Davies: It does, yes, but we are catching up. Rather than having just one intake, we have extended our process, so that we are filling those gaps. It is just taking longer than in a normal year. We will get there on the number of trainees. The standard mitigation for that is to use temporary staff who are clearly qualified, so they understand audit, and we are lucky that we have a fair number of people who return to work for us at our peak periods each year. They are not brand new to NAO work; they are often people who know our systems well and know our audited bodies well.
Q6 Chair: Are they people who have left the workforce or have gone to work elsewhere?
Gareth Davies: Some people choose to work as temporary staff because they are able to pick and choose when they work. Also, obviously, it is a pretty good day rate, so they can work for, say, half the year and cam afford to live that way and pick and choose which contracts they work on. For us, it is a last resort, in the sense that it is not a great use of our resources and it is a quality risk, particularly if it is a temporary auditor who has no knowledge of public sector audit, for example. We try to minimise their use, but if it is a question of having a gap in the team that needs filling, that is the way that we would fill it.
Q7 Dame Meg Hillier: You mentioned a time when you had to bring in outside experts to look at the value of assets. First, how hard is it to find those people? Secondly, can you explain a little more about the new standards? How much can you assure yourself that the Department has done a good job, and how deep are you expected to go? You suggested that discussions were still going on in the profession.
Gareth Davies: I will bring Kate in on this, who can explain how we go about sourcing the expertise. Essentially, we have a framework contract; we can call on experts from it, whether it is for pension fund valuation from actuarial experts or property evaluation from surveyors. Kate, do you want to explain how we go about securing that?
Kate Mathers: Of course. As Gareth said, we have various framework arrangements in place. This is with the major accountancy firms, or for property, with surveyors in well-known property surveying companies. We have put those in place over the years to make sure our teams can access expertise quickly, when they need it and from reputable organisations. We have quite a rigorous process to set up the frameworks in the first place, so that we know the credentials of the people we are working with and so on.
Gareth mentioned earlier some of the challenges raised by the audit quality review—the FRC’s inspections in this area. There is an auditing standard that lays out quite carefully how auditors work with their experts when they are appointed. We follow the requirements of the standard there. We are talking about understanding the capability of the expert, their independence and working with them to carefully scope the work—to set out what is covered and what is not in the work that they do. Also, the standard gives you guidance on how to assess the work of your expert. That is the area where we are looking to tune up. We find the expert clearly has the knowledge, but the question is how we demonstrate that on our audit files. We have made a number of changes over the years so that the teams can access the expertise that they need on a range of subjects, principally around property, pensions and financial instruments. Through the frameworks, we have access to pretty much whatever expertise we need, whether that is on IT audit or in other more niche areas, as and when we need it.
Q8 Dame Meg Hillier: All audits have their challenges, but with pensions you can look at the figures, and it is, relatively speaking, a desk exercise. However, the Public Accounts Committee has looked at property—I am thinking of the Ministry of Defence—and the data on property from Government Departments is poor. It seems to be a huge challenge for the auditor to work out a valuation different from the Department’s. I wonder how you are juggling that.
Kate Mathers: That is absolutely right; there are some big challenges there when you have Government Departments with large, complex estates up and down the country. It is for us, working with those Departments, to explain clearly the audit requirements. I suppose I would say that there is a rising bar of quality requirements set by our regulator, and a definite toughening up of standards. They are setting out more detailed requirements for auditors to follow. Our teams are working with Government Departments and saying, “This is what needs to happen for us to sign off on that part of your accounts.” A partnership approach is required. You will see that our teams go to audited bodies and explain the audit risks and our approach to tackling them. It takes the Department, as well as us, to make that successful.
Q9 Dame Meg Hillier: Finally, Comptroller and Auditor General, does that mean that we might see a qualification of accounts on the basis of poor data on major capital assets?
Gareth Davies: Potentially. There are some examples in the audits that are being signed off for ’21-22. An example is the Environment Agency in the DEFRA group of audits. We have identified problems with the asset records for flood defences. That turned out to be material to those accounts, so you will see a qualification in this year’s accounts on that. That is a good example of where this has been applied in practice.
Q10 Peter Grant: Good morning. First, can I pick up on the comment you made just now? You described a “rising bar” of expectation. That almost seems to be saying that if you keep your audit quality at the same standard, your scores will go down every year, so if you get the same mark every year, it might look to us like a standstill, but it would be a sign of continuing improvement. Is that correct?
Kate Mathers: I think it is very much a case of continuous improvement, and demanding expectations set by an assertive regulator to bring standards up across the profession. That is how we experience that. Something that may have been perfectly acceptable five years ago or three years ago would not be acceptable today, because standards and expectations move on. As I said, there is a general toughening up of expectations from the regulator.
Q11 Peter Grant: Thanks. Mr Davies, you commented on the timing of when the assessment reports come in, and mentioned how late that is in your audit cycle. I understand what you are saying, but the report we have is for audits that you started to plan two years ago or more. First, the whole skill of an audit is that you do not just blindly follow a plan; you adapt it every day, depending on what you find. Secondly, a number of the areas marked as areas for improvement had to do with not the planning of the audit, but the execution of it. Is it possible to put a figure on the extent to which, in the areas marked as areas for improvement, you have made improvements in planning that you could not previously? How much is down to improvements in execution, which you should have made without having to wait for the report?
Gareth Davies: We do dig into that ourselves. As you say, if you are ensuring consistent application of your approach, that removes one set of quality risks, even if you then have to go back and improve your approach overall. We did think about those audits in that frame. How many of these areas represent, to us, a lifting of the bar—and we might not have spotted that ourselves—and how many were execution problems? We think it was half and half. Out of the four audits identified as needing improvements, I do not think we would have said that two did. We were pretty happy with two of them, but now we know that we need to do a bit more, particularly on the point about valuations from experts.
In the other two audits, there were execution issues; it was a matter of consistency of application. That is, we knew how to do it, but did not do it in those two cases.
Q12 Peter Grant: Thank you. You have previously raised the whole question of the timing of the assessment cycle. Though the report from the Financial Reporting Council dates from a few days ago, certainly, when I worked in internal audit, if we found anything significant during an audit, we did not wait until publication of the final report; we told the people responsible immediately. There was also the draft report, which could sometimes take weeks or even months to finalise. During that time, the audited body would know where your thinking was. When did you get an indication from the FRC of the main findings in the report that has just been made public?
Gareth Davies: In some cases, some time ago—a few months ago. These reviews have been done on a rolling basis. Elaine, when would the first results have come in on these reviews—was it February, March?
Elaine Lewis: The very early ones did. It was mainly in the spring and in the summer. As you would expect, as soon as we start to understand what the FRC’s findings are, we issue guidance and training to our auditors on the emerging findings. We do not wait for the final report to come out, no.
Q13 Peter Grant: Okay, thank you. You have mentioned already that one of the issues that keeps coming up is the reliance on the work of experts. Is there an issue of culture, in that finance professionals feel uncomfortable if they are going nose to nose with other financial professionals and saying, “You still haven’t convinced me here”?
Gareth Davies: I think there used to be. Before I arrived, there had been a lot of work under way at the National Audit Office on encouraging a culture of consultation. In other words, it is not a mark of pride for the lead auditor for a Department to feel that they are the expert and can make every judgment themselves.
That culture was already changing to one where it is a mark of strength that you seek expert advice on risky areas and—this is a culture that every firm has been working hard on—“This is the NAO’s audit, not your audit, so you should bring the whole expertise of the NAO to bear on the judgments you are making.” We have some very experienced people who have seen pretty much every situation that you could anticipate, so consulting with those people is bound to result in a better judgment—as is involving external experts in the areas we have already discussed, where that is appropriate. That is now seen as a badge of high performance rather than somehow of weakness. I think the culture has successfully moved on, I would say for the last five years.
Q14 Peter Grant: Realistically, how many areas of specialism can one auditor be good at at the same time?
Gareth Davies: That is a good question, and it is very apt for us. Clearly, we do need people who are experts in financial instruments and, essentially, financial services organisations; several of our audits are in that category now. Some mainstream Government Departments have significant financial instruments in their accounts nowadays. So there is an obvious area where it is possible to be an expert.
We have two approaches to this. One is that we develop sector experts, because it is very important to understand, for example, the complex system of accountability in the national health service as part of the Department of Health group. We develop expertise in these sectoral areas. We also have centres of expertise for cross-cutting topics. In financial audit terms, we have a lead director for pensions valuations, a lead director for property valuations and now a lead director and a growing team on financial instruments. They will obviously have input into audits across the portfolio, rather than expecting every director to be an expert in all those thematic areas.
Q15 Peter Grant: The assessment cycle we are looking at just now is audits that were planned and often carried out in the thick of the pandemic. How much difference did that make to your ability to assess the work of experts, when, very often, you weren’t allowed to meet them?
Gareth Davies: That is an interesting question. Most of that assessment is on qualifications and experience, proven track record, and then being able to challenge their judgments by reference to benchmarks and external information. That was all possible, I would say, during the pandemic. I think generally we were surprised by how effective audit work could be without being able to physically visit the premises, either of the audited bodies or of the experts we were using.
Kate Mathers: I think our teams managed pretty well, actually, under those circumstances. As we are returning to much more face-to-face working, I do think there is a real advantage to that when teams are grappling with quite complicated judgments and so on. Some things are easier done face to face, so I think that will help us in the longer term as well, but I agree with Gareth: I think the teams rose to that challenge very well under the circumstances.
Q16 Peter Grant: You mentioned the reliance on things like qualifications, experience and track record, but isn’t it the case that every wholesale financial calamity that any of us can remember was caused by somebody who, until it all went wrong, had all those attributes? Maybe there were warning signs; maybe there weren’t. What do you do to make sure that you are not dealing with a very highly qualified but also incompetent, negligent or crooked operator?
Gareth Davies: Obviously, it is not as if we are consulting people in areas that we have no expertise or experience. We are pretty good judges, actually, of whether we are getting added-value expertise, or whether we are talking to somebody with extreme views on something, which makes them an outlier and therefore either very useful or potentially questionable. That is where your own experience comes in. Just listening for a predictable view isn’t necessarily a sign of good quality, if somebody is challenging something.
Take, for example, the sudden crisis in pension funds in September or October, which experts did not see coming. That is a good example. Luckily, it is not a financial audit issue at this stage, but auditors’ risk assessments of those pension funds will not have adequately identified that risk, I suspect, despite engaging some of the best experts.
None of these steps in getting assurance are guarantees of high quality. This is where the professional scepticism quality comes in. You have to continually challenge yourself and the audited body. Have they taken the right advice? Does that advice stand up to challenge and scrutiny from people who know what they are talking about? Have we taken on board the challenge from our own experts? You can see why this is such a challenging area. It is very difficult to say, “We’ve got this completely right and we can rely on the views of both our team and the experts advising them,” because events will always prove somebody wrong. It is a question of that process of challenge and making sure we have the best possible people on the case.
Q17 Chair: I would like to bring in Jack Brereton, but can we just for a moment pursue the point you just made about pensions? I was certainly surprised by that. I remember that years ago the Occupational Pensions Regulatory Authority, which the Public Accounts Committee looked at based on a report from the NAO, turned out to have no objectives, and we experienced all the problems of the financial crash. Here we are, 10 or 12 years later—or more—and suddenly, out of the blue, appears a storm that, in a variety of pieces of scenario planning, one might easily have thought might happen if there were a sudden change in interest rates or considerable instability caused by a Government action or other action, but nobody saw it coming.
You said there is no audit issue at the moment, but certainly we have a pensions regulator and that gets audited and can have value for money audits done on it. Is this something that the NAO is planning to look into further? It was a very big surprise to a lot of people. I was staggered when I first came across it, and I kept the FT, which had the Noddy explanation in it, because I found it so difficult to understand how it could have happened. Are you looking at this further?
Gareth Davies: We will have to, because we have pension audits and so understanding pensions’ exposure to this particular risk will be very important now. The issue arises from a perfectly sensible, on the face of it, investment strategy; if you have a closed pension scheme with predictable liabilities, investing your assets so that they match the profile of those liabilities does seem like a sensible plan. The way it was done, using these liability-focused investments, gave rise to this risk in an environment where interest rates were suddenly rising.
Having worked on boards with pension funds of this kind, I don’t remember being advised by any of the investment experts who helped those boards with their plans and who were warning of that particular risk. I think it is another good example of where there is an imperfect understanding of risk, which in our value for money work is coming up across the piece now. Essentially, there is an inadequate understanding of the risk profile faced not just in that situation; we have just seen this on a big scale in the pandemic, for example.
Q18 Chair: But it wasn’t like it hadn’t happened before. We had an example in September 1992, when interest rates rose five percentage points in one day. I lived in Germany at the time. The British Chancellor of the Exchequer wasn’t often on the Berlin evening news, but he was that day. It wasn’t like it wasn’t known that this could happen; it had happened. How did they miss it?
Gareth Davies: I think the issue is the quality of the stress testing. Stress testing of banks, for example, is now much better than it was before the financial crisis, because that experience drove a strong improvement in the stress testing of banks’ capitalisation. I think what has happened here is that people didn’t appreciate the need to stress test pension funds with this kind of investment strategy. Now they do, of course. I am sure we will now see from the regulator stress tests that require your investment strategy to be probed for its exposure to this particular risk.
Q19 Jack Brereton: I want to dig a bit more into some of the high-risk and complex audits relating to financial services. The FRC has again identified an “unacceptable trend” in the quality of these audits. This is not something new. We have seen these issues raised by the FRC for the last seven years. Obviously, we raised these concerns with you last year as well. Do you think your progress in this area has been acceptable?
Gareth Davies: No, because our scores are not where they need to be, so we need to do more. I take it back to the point I started with: the intense action we have been taking in response to the FRC’s findings in the last two years does not show up in these results, for the timing reason I explained earlier. We would be very disappointed if the same findings came out next time around, because we would have had a year more for the improvements we put in place to take effect. This is a continuous process. We are still strengthening our financial instruments area of expertise. Kate might want to say what we have done specifically.
Kate Mathers: Over the last couple of years, we have established our financial instruments centre of expertise, with a director in a leadership role there and a team of staff to start building the knowledge that we need in the NAO. There has been intensive training there to upskill. We also have partnership working in place with some of the firms, which have access to much wider networks of experts in this area than we do. We are bringing training and support in there so that we can upskill to the level that we think we need ourselves.
We have also done a lot of work recently to help our teams understand and assess the risk of these more complex assets—to get their arms around the nature of the risk effectively in the first place—and, thereafter, provide them with more detailed guidance on how to go about executing the audit plan. That is quite an intense area of focus for us, as you would expect.
Gareth Davies: I will just add to that to give you a flavour. Obviously, I see all our audits at the final stage, as we prepare to certify them, and I am able to review the file and see the areas of risk we have worked on and how we have addressed them. We are doing some really good work in this area on financial instruments.
For example, the audits we have just signed off included our first year of audit of Reclaim Fund, which is the body that deals with dormant bank accounts and funnels that money, after appropriate checks and time periods, to good causes. It is a really important bit of the financial architecture now. Our audit of that identified some important risks, challenged very hard on the way some things have been accounted for, resulted in some significant changes in the way those were being dealt with, and added a lot of value to the financial statements in the process.
That is just an example. I see good evidence that our teams understand this area and are raising their game based on the work we have been putting in over the last couple of years. They are giving a good account of themselves in these pretty complex audits.
Q20 Jack Brereton: The FRC has recommended that you should ensure that only those with the appropriate skills and qualifications are allocated to these more challenging and high-risk audits. You said in your response that you would do this where you can. That does not suggest you will do it in all cases. Do you have a shortage of the necessary skills and specialists required to undertake these audits?
Gareth Davies: We are building up the number of our people at every level—directors, audit managers and team members who have the specialist experience we are talking about. We are still increasing those numbers, but I am confident that nobody is in charge of those audits at the moment who does not have the right experience. This is not just the direct members of the audit team; we also have the EQCR process, which is effectively the second partner, to challenge the team on the quality of its judgments at key stages—both at the planning stage and when actually executing the audit. That is a really important bit of adding expertise in here. Making sure that our EQCRs have the right experience to supplement the knowledge of the team is another thing we are doing. I would say we were nearly there on that but, because of the growth of this area of work, we have had to continually increase the numbers involved to keep up with it.
Q21 Jack Brereton: So you are confident that those who you have allocated to those audits are capable of undertaking them to the level of specialist skill required.
Gareth Davies: Yes.
Q22 Jerome Mayhew: If I understand your evidence correctly, you recognise the shortcomings found by the FRC, but you say, essentially, that there is a timing issue—the audits reflect work that was done 18 months or two years earlier—and that, to see the full benefit of the work you have now put in place, we have to wait another 18 months or two years. When the FRC says that you must reassess whether your planned changes, which you have been putting forward as part of the audit transformation programme, are sufficient, is your answer to say, “Calm down. Don’t worry. It’s all in hand. You’re going to see it next year”? Or do you accept the FRC’s challenge to you that, “Look, this isn’t good enough. You need to satisfy us that the work that you are doing really is adequate when the evidence to date suggests that it isn’t”?
Gareth Davies: We definitely accept the challenge. It is one that we apply to ourselves as well. That is why we are still increasing our response. Because of these findings, we are still increasing our focus in areas such as financial instruments—that is the main one identified. The overall point about the reliance on experts has led us to explicitly change our guidance to all of our audit teams this year so that we pick up the point that they have raised, which we had not identified ourselves before. Now that we know about it, we have identified that across the piece.
That is not in any way to say that we are complacent about this, or that it is just a question of waiting. This is a continuous process. I do not think there will ever be a year where we are not enhancing our approach in response to feedback through this process and our own reviews, because it is a continuous improvement cycle.
It is also important to say that none of the audits, even in this set of results, which are not where we need them to be, were deemed “unacceptable”—the FRC’s fourth category. They do use that category for some of the firms they inspected in other areas. While we have areas to improve, they are not “significant improvements” in the language of the FRC. But do we use these results to challenge the way we do it and to improve the approach? Absolutely.
Q23 Jerome Mayhew: Just taking one example, in one of the two audits where they identified recurring findings from the previous year, those included “insufficient testing over probability of default assumptions as well as…significant increase in credit risk…process” and “insufficient procedures performed over the reliability and integrity of data used within the calculation of the expected credit loss.” Don’t ask me to define what that means, but I hope that you guys know what it is. Can you say, “Right, we’ve looked at that and have effected change already,” either planned or in response to this? Can you explain briefly what the change has been? Elaine, you might be the expert on this.
Elaine Lewis: Yes, we have. Each year we produce a quality plan that sets out our priority actions for the year. We are just finalising this year’s. We take on board the FRC’s findings. On those ones that you pulled out there around probability of default and increase in credit risk, we specifically made changes to work programmes, guidance and training to address those, so we have built those into the audits that will be happening now.
Q24 Jerome Mayhew: I think a lot of the longer-term improvement is based around the installation of a new piece of software, which reads as though it tells you what to do next in order not to miss anything. Is that basically what it does?
Gareth Davies: The way we put it is that it makes it easier to do the right thing and more difficult to do the wrong thing. It can never replace auditor expertise, but it is a set of guide rails that keep you—
Jerome Mayhew: It is like airline pilots. They know how to fly the aeroplane but they still have a checklist saying, “Have you turned the engines on?”
Gareth Davies: Yes.
Q25 Jerome Mayhew: Okay, good. What is the timeframe for that to be up and running?
Kate Mathers: Our audit transformation programme is a multi-year programme making some big changes, and it is well under way. Last year we made some quite significant changes to our audit methodology to address some of the findings that are coming out there. This year we introduced a new piece of software for our teams to assess risk in a more granular, nuanced way, because that is absolutely critical at the outset of the audit. We are also building a new audit management system that we should start introducing to teams next year.
Q26 Jerome Mayhew: In ’23?
Kate Mathers: This time next year we expect teams to be planning and starting their audits on the audit management software. This is to Gareth’s point about making our audit approach much more guided—guided workflow—to help auditors get it right first time and understand what the requirements and procedures are.
Q27 Jerome Mayhew: So at this meeting this time next year we can hold you to account that you have got that management process in place.
Gareth Davies: Yes. On the timetable for the new system, the risk assessment element that Kate mentioned is being rolled out from now on for this year, having been successfully piloted last year. We have some pilot teams set up to pilot the audit management software, and the results of that pilot will inform the roll-out to everybody this time next year. Sitting here this time next year, we should be saying that all our teams have now been trained and are ready to use the new software—essentially, the finished package.
The way that the approach has been designed is that, once the software is in use, we can then, every year, add in data analytic tools and plug-in modules to use in the audit as we develop them and as they become available on the market. This is a rapidly changing area of audit technology, and this new software gives us a platform to be keeping up with the rest of the industry for the next 10 years. It is a really important development for us, and we should be able to explain to you the benefits in quality and efficiency of all that investment.
Q28 Chair: Can I return to the question of hard-to-value assets? I take it that this stems from the taxpayer’s unwilling but necessary involvement in propping up much of our banking sector after the financial crash in 2008, and that we still, as taxpayers—through the public purse—own a significant number of shares in banks. That is right, is it?
Gareth Davies: It is where some of this originated. It is now a very wide class of assets across Government, actually. Others might want to add to this, but we are talking here about anything for which you cannot read a stock price in the Financial Times in the morning. A lot of what the Government has invested in is in that category—almost by definition, because it really should not just be investing in mainstream companies. It is usually unusual investments, either for national security reasons or for strategic industrial reasons.
Q29 Chair: So we are talking about hard-to-value assets that are not complex financial instruments. We are talking about things that have other risks associated with them.
Gareth Davies: Unquoted companies, our contributions to international partnerships—
Q30 Chair: Or development finance institutions.
Gareth Davies: Exactly.
Q31 Chair: Can I just for the moment focus on the complex financial instruments? I think I am right in saying that you have a centre for expertise on financial instruments specifically, because of the added complexity. Obviously, you have had to do it because of the Government’s involvement in owning banks—financial instruments come with that—but have things developed to the point that that will be a permanent feature of the scene? Is it not the case that eventually, when we get to sell the shares back to the private market, that need for expertise will disappear?
Gareth Davies: No.
Q32 Chair: Why not?
Gareth Davies: Well, for example, you now have the loan book for bounce back loans, which is a 10-year scheme, and the expected credit loss calculation has to be updated each year in the BEIS accounts. That is a good example of something that was not there before.
Q33 Chair: And presumably also there are times when the Government wake up in the morning with another idea for a bank—whether it is the Green Investment Bank, the British Business Bank or CDC, now known as British International Investment, which has many of the characteristics of a bank. Those things are not going away either. Is this basically an ongoing area for investment, and does the FRC think you need to invest more in this?
Gareth Davies: They do, and we are. We think we will have to continue to grow the number of experts. In response to Mr Brereton’s question, we will continue to need to expand the number of auditors specialising in this area, and that includes hiring people directly from firms that have that expertise, as well as growing our own and increasing the in-house centre of expertise that we have built up in this area. We are also letting a specific framework contract just for financial instrument expertise, because it has become such a dominant bit of our requirement for external help that it needs its own framework contract. All those are signs that this will be with us for a long time.
Q34 Chair: You say “hiring from the firms”; will we see people being hired from the major investment banks, with telephone-number salaries? They made the mess, and you are now hiring them to help you audit how it is cleared up.
Gareth Davies: Not at the NAO, no.
Q35 Chair: So what are you talking about? Hiring compliance people from major investment banks?
Gareth Davies: No, hiring auditors—people who have experience of auditing complex financial instruments.
Chair: Internally. Okay, good.
Q36 Peter Grant: The issue of valuation and revaluation, especially of intangible assets, leads on nicely to the subject of testing journal entries. A good example of where journal entries are required is where you revalue an asset. There is not a cash transaction, but you may well have to make an adjustment to the books. How did you react when you realised that, yet again, the FRC was going to pick this up as an area for improvement?
Gareth Davies: I think this is a good example of the improvement cycle working its way through. You may remember that last year, we talked about the journals as a key finding of the reviews. It led to some of the scores being below the line, as we would say—requiring improvement. This year, while the FRC does not think we have quite finished that off yet, none of the issues they identified on journal testing pushed our score below the line, so that is an improvement on last year. We do not want to see the issue raised at all in this report. As we have finished our push on training and on updating work programmes, so that our teams are completely clear about what is required, we are hopeful that the issue will not appear at all in next year’s report. Obviously, the proof of the pudding will be in the eating, but I think that is a fair reflection, Elaine, of the cycle we have been on, on journals.
Elaine Lewis indicated assent.
Q37 Peter Grant: In the most recent report, the FRC reported that in seven audits, the audit team did not sufficiently justify why certain journals were excluded from further testing, including journals that met the high-risk criteria that the audit team had set. It also said that in two of the audits, the audit team did not adequately evidence how the risk criteria that they used to select journals for testing matched with the risks identified earlier. Those areas for improvement were not sufficient to bring your overall score below the line, to use your term, but they are still quite significant grounds for improvement. In your response, you said that during ’21-22—after you had done the audit work, but before the FRC had reported on it—you had provided mandatory in-depth training to all your audit staff about what they needed to do in testing journals. How confident are you that that training was sufficient, and most importantly, has it changed the way that your staff do these tests?
Elaine Lewis: The findings that you highlight are important, and we definitely want to deal with them. They were around the way that we were evidencing our judgments. From discussions with the FRC, we know that the reason why they are not grade-driving points is that the judgments we made were satisfactory, but we had not properly evidenced them on the file, which of course we should have. The training was very much designed with the prior findings in mind, and to make sure that we address exactly those points. I have done the training, and it definitely does cover all those.
As for checking whether that has been done, we have a programme of hot reviews on audits, in which we check, while audits are ongoing, that we are complying with our requirements and our guidance. Part of that included checking that people were following the training in the audits for ’21-22.
Q38 Peter Grant: How often do you plan to repeat that training?
Elaine Lewis: The training is ongoing. People who did not do it last year because they have only just joined us will do a repeat of it, but we will probably keep considering and updating it, not necessarily every year but on a cycle, so that we make sure that we keep up to speed on this, because it will always be important.
Q39 Peter Grant: Have you also built more rigorous review processes into the NAO, so that when an audit is passed up for review, the most senior staff doing the reviews concentrate on a particular area?
Elaine Lewis: Absolutely. The training also covered review, and was for staff at all levels in the organisation. We have also improved the documentation package that we use to record our work on journals, to make it very clear how these judgments need to be documented; that also makes this easier to review.
Q40 Jack Brereton: The NAO’s internal quality monitoring cycle had not concluded when the FRC’s report was conducted. Could you explain what caused that delay, and why it was later than in previous years?
Gareth Davies: This is essentially the impact of the pandemic-induced delays in our core audit work. The people who do our internal reviews are audit managers and others released from audits to do them. Where we had delays in completing the main audits, that has had a knock-on effect on the cold review programme—our internal review programme. Do you have anything to add about the causes, Elaine?
Elaine Lewis: Yes, those are the causes, but it is absolutely true that we want to get those reviews done as early as possible for exactly the same reasons that we want to get the FRC reviews done.
Q41 Jack Brereton: So you are confident that we will not have these timeliness issues in future.
Gareth Davies: It is a joint effort between us and the finance teams in Departments, but we are successfully dragging the audit timetables back to where they should be, to their pre-pandemic position. Four more big Departments were certified before the summer recess this year, including some very big, complex audits, such as HMRC and the MOD. That is a better position for the coming year. We are not quite there yet on some important audits, such as the Department of Health and Social Care, which, for obvious reasons, will be the last to get back to a pre-recess timetable.
Q42 Jack Brereton: This is mainly for Elaine. If you compare the results of your internal monitoring with the FRC’s report, there is quite a difference. If you look at the number of “good” or “limited improvements required” ratings, the two together account for 75% of audits, according to your internal monitoring results. The figure is only 56% in the FRC report. What is the reason for this difference? Do you think your report is telling a different story?
Gareth Davies: That is a really important question that we get into in a lot of detail. We have discussed that with the audit quality board. Janet might want to add to what I am saying. I think there are two main reasons for that difference. There is not always a difference, actually, and often they have been quite close together. First, the samples are picked on a different basis. We make sure we cover all our teams, whereas the FRC sample is its own risk-based sample. I think they pick audits they think will be tricky, so it is a riskier sample.
Q43 Jack Brereton: There is quite a difference between 75% and 56%, isn’t there?
Gareth Davies: Yes. I will address the second one, but Elaine, do you want to pick that up?
Elaine Lewis: I was just going to say that I think that was based on the results at the point when the FRC reported. We have concluded our programme since then. We have 65% above the line in our final out-turn, which is still obviously higher than the FRC’s 56%, but it is not such a big gap.
Gareth Davies: There is another reason why you sometimes get a gap, usually in this direction. By definition, as we were discussing earlier, where the FRC says, “A higher standard is now required on this,” we will not have picked that up in our internal reviews in that year, but we will the following year, when we have built that higher expectation into our approach. Typically, you will see this difference, and then our methodology catches up with theirs and they should be back together again. Occasionally, you do see this gap.
Q44 Chair: May I bring in Janet Eilbeck? You have been a non-executive director of the NAO for six years. You are, to use the American parlance for certain governors, term-limited, and you are coming to the end of your service as a non-executive director. You chair the audit quality board. Can you reflect for us on the changes that you have seen, and say what more you think is needed?
Janet Eilbeck: First, I should correct that; my second term as a non-executive finished in October.
Chair: Oh, I am sorry. But you are still chairing the audit quality board.
Janet Eilbeck: I am still chairing the audit quality board as an independent, that is right.
Chair: Well, I stand corrected, and the record will show it.
Janet Eilbeck: Thank you. Yes, I have seen a lot of change. The C&AG talked about the improved consultation willingness, and the openness to conceding when something in your audit has not gone quite as you wanted, so you do not get the score that you wanted. In particular, there has been a big change in consulting on difficult matters. Some of the training programmes that encourage people to talk about more difficult experiences and how they overcame those are new. I have seen a big change.
The use of experts is definitely different and more advanced than it was. I have seen a really solid progression there. The NAO has always had really capable people, and that is still the case, but I think management systems, control over aspects of quality and methodology, and in particular the people aspects of how audits are put together and delivered have all moved on considerably.
Q45 Chair: When you say consulting, you mean consulting internally with colleagues.
Janet Eilbeck: Internally, yes.
Q46 Chair: That surprised me, because I used to be a legal journalist many years ago, as well as a financial journalist, and in professional services firms, including law firms—I worked as a consultant for one—the first thing you did when you had a problem was chew the cud with colleagues. By the way, I also represented the management consultancy industry; I have a fairly chequered past. One of the points that the industry used to sell itself on was that you were not buying the expertise of one individual; you were buying a consultancy—a collegiate entity where people talked to each other.
We have always understood that the NAO has many similarities to the professional firms, and that, in some ways, it is like a consulting firm. We heard from a previous C&AG many years ago about the added value from insight that could be offered. Why is this change happening now? Why was that not embedded many years ago?
Janet Eilbeck: I think it is a progression that has been visible. The C&AG talked about colleagues who were the expert in their department, and they felt, in the past, that there was no need to consult. However, now, I think there is the same recognition in the NAO of the value of consulting that we would expect to see in the firms. I used to be in one of the firms, and that was part of the familiar culture to me. I see that now in the NAO. It is now routine for the technical department to consult on a technical matter, and for a number of people to get together on a review panel to thrash out a difficult technical judgment.
Chair: That is very good to hear.
Q47 Anthony Browne: My question is about the comparison with the audits in the private sector, particularly those overseen by the FRC and ICAEW. I have comparisons here from the past three years. In 2019-20, more than 70% of your audits were categorised as “good” or “limited improvements required”, which was higher than those looked at by the FRC or the ICAEW. If you dial forwards two years, the latest data shows that the private sector firms have made dramatic improvements, and are now way ahead of you on audits categorised as “good”.
If you look at the latest year—this is FRC only, as opposed to ICAEW—Grant Thornton has a remarkable track record, with 100% of audits categorised as “good”. KPMG, PwC and Deloitte had over 80% categorised as “good” or “limited improvements required”, and you are down below 60%, at the bottom of the pack. Is the truth not that you have been falling behind the private sector on audit quality?
Gareth Davies: We are certainly behind in this year’s results, which you have pointed to. In the previous two years, we were broadly in the pack of the big four, which is the comparison that we make, and well above the mid-tier in those years. As you have hopefully detected from us, there is a determination to ensure that we are not below the standard of the rest of the market. I think that Parliament should expect the best available auditor for these important accounts, so that is where we need to get to. We have described the work that we are doing to ensure that we return to that.
We invited the chief executive of the FRC to address our senior leaders—all the audit directors, and other senior people in the organisation—back in October. He gave us an excellent summary of the FRC’s findings at the big firms, and compared them to ours. His message on that was, “These are the same issues that we are picking up, but you are a year or two behind those firms.”
He also made the point that the big four have collectively invested £4 billion in audit quality in response to the big failures in audit quality in the commercial market, which we all know about. He said, “Everyone understands that the NAO will not be able to invest at that scale; the challenge is, can you invest proportionately, for your size and the complexity of the work, to address the same risks?” Those are the things that we have been talking about, around financial instruments, reliance on experts, and so on.
That was a very useful bit of perspective for us. However, our conclusion, having heard his views and discussed the matter ourselves, was that it is an achievable aim for us to score in the same way as the firms, even though we don’t have access to the billions that they have invested. That is because, first of all, we have a very focused market for our work; we need to be experts in public sector audit. The range of issues you need to be good at to do that is now broader, as we have discussed this morning; it includes financial instruments and other things. However, it is achievable. We have the investment in our audit transformation programme, which this Commission approved a couple of years ago as being the right one, and we are on track to implement that. We believe that that will allow us to get the scores necessary to be in the pack with the biggest firms. That is our aim; I think we have the right plans for achieving it, and we will not be happy until we do.
Q48 Anthony Browne: You mentioned the importance of investment, and what you said is obviously true, but are there any other reasons why the NAO dropped behind the private sector in audit quality?
Gareth Davies: Every audit firm has had to audit organisations affected by the pandemic, but not all of them have had to audit organisations that have been in the lead in responding to the pandemic. We had huge issues to respond to as an organisation with finite resources. I am not putting that up as a reason why our quality might have suffered. I do not think it has. In fact, in its own way, it has been a real confidence-booster to our audit staff, because I think they have performed excellently. If you read the audit report on the Department of Health and Social Care for 2021—the year that we are looking at here—I think it is a model of clarity in dealing with really tricky issues on assurance and public confidence in spending, and is a real service to the public interest.
Alongside these scores—I think I have demonstrated that we are taking them very seriously—there is a lot for us to be very confident about in the quality of the work that our teams are doing. There is a very interesting question about building the confidence of the organisation—confidence that we are doing the right things to achieve high quality. The feeling in the audit teams at the moment is that we have done an incredibly good job through the pandemic in difficult circumstances. There have been some very high-profile audit findings and reports that have made a big difference, I think, and we have responded well to all the logistical pressures of lockdowns and so on.
Why do we not feel that we are keeping up with the quality in the wider market? Clearly, we need the scores to turn round, so that people see that we are doing the right thing. They should be proud of the quality of the work that they are doing, and it should stand up to the best in the rest of the market. That is a big organisational imperative for us.
Q49 Anthony Browne: You mentioned various things there and earlier that are needed to raise audit quality. One of the things the FRC emphasises across the audit sector—both you and the private sector—to raise audit quality is that you need “a healthy culture…that encourages challenge and professional scepticism”. Do you think there are cultural issues across the audit sector about “challenge and professional scepticism”, and if so, what are you doing in the NAO to try and promote those things, assuming that you are?
Gareth Davies: It is a general issue. We have our own specific version of it, because one natural advantage that we at the NAO start with is that we do not have the threats to ethical standards that occur in the commercial practices. Our auditors do not have to fend off the consultancy teams of the firm that would prefer to be selling high-value consultancy to the client. A former colleague of ours who joined us from the private sector described the NAO as “the auditor’s paradise”, because you are not looking over your shoulder at being reappointed at the next AGM, and you do not have to fend off other parts of the business that are trying to impinge on your audit relationship. That is a big strength that we have. There is a real culture of public interest around the quality of the audit, which is really important.
On the other hand, we deal with a narrow segment of the economy, so we do not get the kind of breadth of sectors that the firms benefit from. In some areas, we have to import more of our expertise than they—the firms—do, when they are big enough to do it themselves.
As for maintaining the right culture of “scepticism”, to be usefully sceptical you need to be very well informed and you need to understand the issues you are grappling with. That takes us all the way back to the issues of specialist expertise and advice. That is the biggest challenge we have in making sure that we are punching our weight with the large firms.
Q50 Anthony Browne: I don’t want to put words in your mouth, but are you basically saying that in your situation—your “auditor’s paradise”—it is easier to encourage a culture of scepticism than it is in a private firm, where you are worrying about being reappointed?
Gareth Davies: It is easier to focus on the public interest, I would say, rather than on the commercial interests of the firm that you are part of.
Q51 Anthony Browne: Does anybody else have any comments about the standards of the NAO compared with the private sector? You don’t have to say something.
Gareth Davies: I think that we are healthily competitive, as well. It is pretty painful to be below the pack this year, in a way that we weren’t for the last two years. The organisation is proud of the quality of its work and the impact that it has in the public interest. This is a very strong spur to everybody to be back where we should be.
Q52 Chair: There is, as we have discussed previously, a continuing covid-related backlog of issues that you have to deal with in some of your audits. From your previous answer, it sounds as if you don’t think that has reduced your bandwidth to be able to deal with audit quality—if anything, you are saying the reverse.
Gareth Davies: It has clearly put pressure on the delivery capacity of the teams, which we work very hard to maintain. Auditors have had to deal with situations that were in no textbooks, so like everybody else we have learned a huge amount in the past two or three years. People have seen the value there.
A very simple example would be the levels of public concern around procurement in the pandemic, which led us to a very focused bit of related parties audit work in the DHSC accounts. If you look at the 2021 statement of accounts, you’ll see very extensive disclosures around that, which were driven by the challenge presented by the auditor in that case. That is just a simple example of the principle of scepticism about financial statements, particularly in Government, needing to inform people in the public interest. That shines through in that example. If you read the rest of the audit report on those accounts, it sets out the depth of work that was necessary for that team.
The teams who have been at the frontline of those responses feel that they have certainly had to flex their audit muscles and respond to an incredibly unusual situation, and they have done that well. In that sense, it has been a real boost to the culture of quality in the organisation. Clearly, we also need to get the nuts and bolts of investment valuations and so on right as part of that.
Q53 Chair: To what extent, if ever, do you turn to Government and say, “You’re doing things that are too complex to account for or to audit”? One remembers vividly the former Comptroller and Auditor General, the late Sir John Bourn, disclaiming the accounts of the Home Office 15 or 16 years ago because there was no information with which to form an audit opinion. I don’t think that was due to a complex financial instrument; it was due to sheer administrative meltdown. We have seen the accounts of the DWP qualified for many years because of the proportion of error that is seen as justifiable, to the point where one might say to the permanent secretary and the Secretary of State, “You’re trying to drive through policies that are unauditable or unaccountable for,” and the permanent secretary ought permanently to be asking for a qualification, but that doesn’t always seem to happen. Can you just unpack that dynamic and that tension? It must exist in a lot of different places.
Gareth Davies: It does. I think there are a couple of issues there. We have come close to a similar situation to the one you describe at the Home Office from some years ago, particularly with the impact of the pandemic on some organisations in Government groups. We have not actually had to disclaim any of our opinions yet, but it has been close in a couple of cases. The issue has not gone away. It is perhaps more excusable in a national emergency than just as a failure of administration; even so, it still needs addressing. One of our messages is that the restoration of high standards of accounting—both timeliness and quality—is still a big issue for lots of parts of Government coming out of the pandemic.
In your DWP example, essentially the level of fraud and error detected in the systems being operated for paying different kinds of benefit remains above our materiality levels. The risk there is that it becomes seen as business as usual, instead of something to be worried about. The big advantage we have with our work is that as well as our technical opinion on the accounts, which follows the professional standards in wording and format, we have the opportunity to issue a report on the accounts. In the DWP case, that delves into this issue of the level of fraud and, particularly, the trends. The worrying information from the most recent DWP audit report is that it shows the highest level of recorded fraud of the universal credit system so far. In other words, it is heading in the wrong direction.
Q54 Chair: There is still a qualification against their accounts? Can you remind us of when their accounts were last unqualified?
Gareth Davies: I think it was 33 years ago.
Chair: I was going to say 1988. That would make it 34 years. It sounds like business as usual to me.
Gareth Davies: Clearly the answer is not to shut off the universal credit system. We are working with the DWP on its investment in improving control over that system, so between us we can chart whether the improvement of controls is having the desired effect of reducing fraud and error. Everyone accepts that it is not going to be zero, but at what stage do we think the investment is sufficient and at an acceptable level in materiality terms? That is obviously a judgment. You cannot read that number off any of the standards.
We were developing what I thought was a promising approach to that; then the pandemic hit and of course the Department had other priorities. We are now getting back to that agenda, where if the Department can show us the improvements it is making, the increasing degree of control and therefore the reductions in fraud and error, then at an appropriate time when the levels have come down sufficiently, we will be able to lift that qualification even lower. Clearly at that point there would still be a level left, but you could deem it to be the economically uncontrollable amount. To keep pouring money into the controls eventually becomes uneconomic.
Q55 Chair: You said that the audit quality issues had not suffered from the pandemic, in the sense that you were actually rather proud of how your staff had responded to it and they had provided some very good insight in difficult conditions. Do you think the training experience of new members of staff suffered because of the pandemic?
Gareth Davies: Yes, I do—not the formal training, because we were able to deliver all our internal courses and our trainees were able to take their professional courses and exams very satisfactorily actually. More important—I think we have discussed this before—is the on-the-job training that is at least 50%, if not more, of the value you need as a trainee. You need to be working alongside more experienced people to see how they deal with difficult situations. We have all, I am sure, drawn conclusions from people working around us about who we want to emulate and who we don’t. You can get a fraction of that working remotely, but only a fraction. It is something we have been very keen to get back as quickly as possible.
Q56 Chair: I want to bring in Peter Grant in a moment, but you have just prompted one more question from me on the appropriate balance between on-site and remote auditing. I remember in a Commission hearing some time ago Peter Grant discussing an auditor sitting quietly in the middle of a room, keeping his mouth closed and eyes and ears open, noticing things you wouldn’t notice if you were just auditing by computer—even if you are doing a much bigger “sample” than previous technology would have allowed. Do you think you have found the appropriate balance between remote and on-site auditing?
Gareth Davies: I think we are getting there, yes. We have had a long period since February or March when the restrictions were lifted. We used that time to get back to a balance we think is necessary. It would be interesting to see what the long-term effects of the pandemic are. We used to send teams to sit for weeks on end at the Ministry of Defence and do the field work. We now take roughly half the time on site we took before, and we think we are getting all the value from being there in that time. It is obviously much cheaper for us to do that too, so it is an efficient way of doing it.
I think the experience of the pandemic will inspire some long-term changes in the way we conduct the audits, but you need face-to-face time and to be part of the office environment you are auditing. Crucially, you build up relationships with the people who are making the judgments in the accounts—the finance team and the audit committee, for example. Our experience of the first and maybe the second year of the pandemic was that you were trading on existing relationships, which is okay, but they quickly deteriorate.
Chair: And people move on.
Gareth Davies: If there is any change, you do not have that relationship with the new people. So yes, we are insisting on face-to-face meetings to set up the work, to sign off the key stages and to meet the audit committee.
Q57 Chair: I know the NAO has done a lot of work in the past on Government estates and the amount of real estate that is taken up, particularly if people are working from home 20% of the time. Of course, the amount of time people spend at home has risen since the pandemic. In winter, you can see very easily at 4 o’clock, when it is dark, which buildings are in active use. In recent months, there have been any number of Government offices where people appear not to be there when one walks past. The offices appear to be almost closed. That presumably has implications for how much space they need.
Gareth Davies: I think this is economy wide, by the way; it is not just a public sector issue. A few weeks ago, I was at an event at one of the big insurance company headquarters in the City. It was a Thursday evening event, and the chief executive ruefully remarked, “If you came in tomorrow, this place would be empty.” It is not just a Government issue. We have just published a report giving an updated view on the value for money of the Government property strategy and the progress it has been making. That is a very recent report, and there was a PAC hearing on the back of it.
There was certainly a risk, a year or so ago, of organisations concluding prematurely that they would work remotely for a significant amount of time, dispensing with a lot of office space and boxing themselves into a strategy that they might now be regretting. I detect generally a shift back to having people in the office, so you have to be careful. At the NAO, I was very clear that we were in no position to make a long-term decision about our property in the heat of the pandemic, and that we needed to wait and see. I am glad that we did that, actually, because we are now much clearer that we are not a remote working organisation. We need to have people together, and we need the space to do that effectively.
But you are right: there is bound to be a shift through all of this. Along with the many other factors that the Government have to take into account, such as their carbon reduction strategy, they have to ask how they will find themselves with an efficient property portfolio that is fit for purpose for the way people are working for the next few years. We will follow up our work on that across Government to keep the pressure on.
Chair: Thank you. Let me bring in Peter Grant to ask a final question on audit quality.
Q58 Peter Grant: A huge part of your financial audit work is auditing the accounts of big Government Departments, all of which have the same deadline, which traditionally is the start of the summer recess. You have previously said that such a big concentration of work at the same time is an issue for you and that, at times, you have thought that it threatened to undermine the quality of the audit work. Is that still something that concerns you?
Gareth Davies: Yes. It would be nice, actually, to have the problem of getting back to that big summer peak, with all the big Departments signed off by the end of July. I think that time limit is important for proper accountability. To peer over the precipice of what it is like if you let that slip, you have just to look at local government at the moment. The timely delivery audits there is in a very poor state; in many cases, there are two or three years outstanding. This is the broken window theory, isn’t it? If you allow a small slippage, it becomes a big problem very easily, so we need to get back to that timetable. We are working very hard to do that, as I say, with good progress this year.
As you say, that brings with it the resourcing challenge of a large peak in our final audit work. The answer to that is to bring as much forward as possible, through good planning and interim work—what are often called month nine audit processes—so that you get a lot of the assurance you need before the end of the year and then you can top it up once the year end arrives. The use of digital audit techniques is designed to make more of that, with more of a controls-based assurance approach where that is appropriate. A big part of the agenda there is pulling more of the work forward and then making sure that our resources are as flexible and available as possible at what will still, inevitably, be a peak period, so flexible working arrangements and the ability to have time off in lieu in the summer when the deadlines have passed and been achieved. There are some pretty well established routines at the NAO for achieving that.
Before the pandemic, when those peaks were an annual occurrence, teams would often speculate about how much better it would be if we could spread some of the work into the autumn and take off some of the pressure. Now they have had to do that, they are less keen in general because the work never ends. For example, the team that has done so well in auditing the Department of Health and Social Care is still completing the ’22 audit and will immediately have to swing into planning next year’s work. It might be hard work in the summer peak, but you do get a break and then you come with a clear head to planning next year’s work. I think people have recognised the desirability of that; it just means a lot of hard work in the months from April to July.
Q59 Peter Grant: Thanks. You referred clearly to what has gone wrong in local authority audit—so many councils have got further and further behind—but is that the same as saying that within, effectively, a group of companies, or a group of Government Departments, it’s not that you allow the Department of Health and Social Care audit timetable to slip, but you change the timetable? You say, “As of now, your accounts have to be audited by the end of September, and we’ll hold you to that.” Does that affect public accountability so much?
The Public Accounts Committee is one of the main ways in which your work on the audit of these financial statements is brought to the attention of Parliament and the public. It takes us a year to get through all the big Departments that we want to do, and we maybe do sessions on a quarter of the audits that you do. Is it the case that having the audit completion deadlines staggered weakens public accountability at all, or does that simply recognise the fact that, if you present 20 or 30 different sets of accounts to Parliament within two days, nobody—not even members of the Public Accounts Committee—is going to read them all before Christmas anyway?
Gareth Davies: I take the point about staggering allowing for them to be taken in bite-sized chunks. The problem is that if you are one of the Departments that has been staggered into autumn, for example, and, say, the audit identifies a material issue with the previous year’s accounts and that is relevant to your current financial position, the sooner you know about that, the better. There is a basic rule that a timely audit is a more useful audit. That is not to be taken lightly. Of course, the fact that the audit is finished by July doesn’t mean that you can’t have a very useful staggered set of sessions on the audited accounts through autumn and winter. I would still rather have timely audits that are available to be used by the Committee when it is able to fit them in, rather than the Committee’s work programme being driven by our audit timetable, which I think is the wrong way round.
Q60 Chair: Presumably, it is also true, because you have regular consultations and discussions with the Public Accounts Committee, that if you have a set of audited accounts for each Department that are done on time, you can in your discussions draw to the attention of the Committee whether there is, for example, a material issue that has arisen in the Foreign and Commonwealth Office accounts, which it did, or whether a material issue has arisen in the Revenue and Customs prosecutions office, which it did. You can have that discussion and make sure that over a period of months, anything you think probably deserves the extra attention of Parliament can get it.
Gareth Davies: Yes. That is exactly how the process works.
Chair: Good. Well, I am very old-fashioned, but that’s how it always used to work. I am glad to hear that you think that too. That concludes today’s evidence session. Thank you all very much for coming. The Commission will now deliberate in private.